Getting your research article published Mark H Ebell

Getting your research article published Mark H. Ebell, M. D. , M. S. Deputy Editor, American Family Physician Former Editor, Journal of Family Practice Professor, Office of the Provost University of Georgia, Athens, GA

Most important advice n Submit your work to the right journal! n Journals focus on: q q q n Clinical research Bench research Health services research Systematic reviews and meta-analyses (+/-) No research at all Think about who you are trying to reach…

Follow the author guidelines! n n n Go to Web for latest guidelines Print them Read them before starting Contact editorial office if questions Read them again before submitting your final manuscript, use any checklists provided

Common formatting errors n n n Failure to provide complete title page Failure to use hard page breaks Failure to double-space and use adequate margins

n Step by step through each section of a research article…

Introduction n n 1 -3 paragraphs Tell the reader: q q q n n What is known What is missing What you are going to add I can usually cut the first paragraph; don’t need detailed exposition on why hypertension or hyperlipidemia are bad Close with simple statement of your study question

Methods n n n Tell us enough so someone could repeat your study and so we can critically appraise it (Was allocation concealed? How was randomization accomplished? Definitions? ) Don’t necessarily repeat what is self-evident from Results, I. e. “We recorded age, gender, marital status, race…” Details about instruments, extensive calculations, etc can be presented online in an Appendix.

Results n n n Don’t repeat what is in tables – just hit the high points of each table or figure. Generally 4 to 8 figures and tables Put tables at the end of the word processor file, not as separate spreadsheets.

Tables n n Do NOT use tabs and spaces in a table. EVER. It is impossible to reformat. Use the built-in Table creation feature in MS Word: q q q Table | Insert | Table Tell it how many rows and columns Can always change rows and columns later.

Tables (how not to…) Outcome Control Mortality 3. 4% Length of stay 4. 3 days Symptom Score Treatment 4. 5% 3. 2 days 3. 2 9. 3

Tables (nice!) Outcome Treatment Control Mortality 4. 5% 3. 4% Length of stay 3. 2 days 4. 3 days Symptom score 3. 2 9. 3

Figures n n Do not embed figures in the manuscript Do not create figures using the MS Word drawing tools Do create them using a Paint or Paintshop program Do use photos and graphics of very high resolution

Figures n Carefully review journal’s requirements n Am Fam Phys q q q JPG or TIFF file MINIMUM 250 kb or larger, 600 x 800 pixels minimum –ideally bigger than that Submit each graphic (figure/photo/drawing) as separate graphics file

Don’t over use figures!

Discussion n n BRIEFLY summarize key findings Put them into context – how does this add to the existing knowledge base? Acknowledge any limitations Point the way for future research

References n Use proper format: q q q n Authors. Title. Journal Year; Volume: Page range. Ebell MH, Barry HC. POEMs in the medical literature. J Fam Pract 1998; 43(2): 341 -4. For Web citations include URL and date last cited. See author guide at www. aafp. org/afp for details

Overall – be concise! n n n Journal pages are limited and expensive Readers are busy and want more concise articles 2000 words is a good target for a typical research article (or review article for that matter) Don’t try to do too much – consider splitting long article into two sections Read BMJ and Lancet for examples of how to present research concisely.

Other issues which lower priority n Highly selected population q q n Confirming that something we already know in a more typical US population also applies in a particular group (I. e. Hmong, Hispanic population) Will not be of interest to most readers, and will not change practice Wrong study design q q Don’t let the dataset drive the question Think about criteria for good study about diagnosis, prognosis, or treatment – can you achieve them?

Other issues which lower priority n Fatal flaws q q n n No control group Poor reference standard Unclear definitions Etc. Uninteresting question DOEs rather than POEMs We have too many surveys in our literature! Low priority doesn’t mean bad or worthless, but a typical research journal (Ann Fam Med) only publishes 80 to 100 studies/year!

Summary Higher priority Lower priority Good question or topic So what? Patient-oriented outcomes Disease-oriented outcomes Concise Longwinded Relevant to clinicians Academic interest only

Summary Good! Avoid… Simple sentences, active voice Stilted, pedantic, academicese Practice changers So what? Clear organization Muddled thinking Clear, simple tables Useless figures

Summary Good! Paying attention to Author Guidelines Avoid… Making me remind you to look at the #$^^%#^%^ Author Guidelines Entire article (except Paper submission or figures in a single file 10 different files Letter accompanying Making me guess a revision telling me how you responded what you did to the reviewers

Thanks! Questions?